36 replies to this topic

[randfish]

I've been getting into some unfamiliar territory recently with a few of our older clients, as well as some new ones. Although I've run up against many Google penalties in the past, I'm seeing some odd new trends that I would love additional insight and input on.

The first situation deals with a site that has been ranking very well in the SERPs for the last 4-5 years. Recently, however, a number of pages targeting specific terms were dropped into Google's supplemental results, then pulled into the "no cache, no title, no description" type listings. This coincided with a dopr in rankings for all of these pages, to where one must click the "repeat the search with the omitted results included" link at the end of 1000 results in order to see the pages listed. The same holds true for the site:url.com command - only when you click that link or use the &filter=0 in the URL do the pages appear.

I've reviewed the site's inbound links, the pages themselves and the rankings for the other pages (which seem, oddly, unaffected at all) - I see no signs of anything untoward. These are static ".htm" pages that have been unranked and unlisted for well over 3 weeks now.

My concern is that because we continue to see pages being delisted and unranked, there's some sort of funniness (not sure I want to say "penalty" because it's such an overly abused term) affecting the site. The pages that drop out appear to be those for the more highly competitive kw terms - pages that don't specifically target or mention tough keyword phrases haven't been touched.

Any ideas? Anyone seen simliiar stuff?

The second piece of the puzzle I'm dealing with is on another site entirely. The site has had positive rankings for about 1 year in a mid-level competitive market. Recently, however, for one specific two word phrase and some variations on that phrase, the site's rankings have dropped from #2-3 to #141... The remainder of the rankings remain consistent, and it does not appear that any other sites that were ranking well for this single phrase have dropped in rankings... I was hoping to find a link that was lost, but the links have actually been growing steadily and organically, although that dropped off when the rankings fell. We've got access to some very spiffy tools for link and site investigation, but I'm just not seeing the data that would produce this kind of result (particularly when other kws the page is targeting are rising or remaining steady at the top).

I know it's rare nowadays to talk specifically in forums about rankings and results in such detail, so I thought I'd bring it back a bit. I'd love to hear your shared experiences and perhaps find a common cause that might be creating these results.

Edited by randfish, 11 January 2006 - 03:08 PM.

[bragadocchio]

Just a quick thought to offer at this point.

Have you checked to see if there are other sites online that are using the same content as those pages? What you've described I've seen when results are being filtered for duplicate content.

Of course, there may be a number of reasons, but that is one that I would check first.

[Nadir]

Did you try to contact Threadwatch and see what they think about what happened to their site?

It looks like the same thing happened to them as you posted it on your blog. This is the weirdest think I've ever seen in my SEO life! It says 2 results,but if you click on omitted results, the pages look good. It could be explained if pages were similar, but in that case, I dont' get it...

[Wit]

Hmmm, a "big-dough-keywords" penalty (oops I mentioned the p-word) that's even heavier than the sandbox????!?

Would put it past Google. Same with using the Big Daddy stuff to creative a positive-looking distraction for what is/was in fact a rather negative filter.

</tinfoil>

Quick question though. Are these websites "official"? I mean are they like keyword1-keyword2.com or are they like cocacola.com, with proper business info and credits and brand-awareness and all that cr@p???


Also note that the dc's are dancing....

Edited by Wit_, 11 January 2006 - 03:24 PM.

[Ruud]

I'm playing with a similar situation, yes. My gut feeling is that it has to do with the internal links and anchor text on those pages. On one page I've been experimenting with going wider with the anchor text and that page has been moving forward a bit. On this site too the symptom is that the specific phrase is being demoted. The strange thing, at least to me, is that for the more general phrase Google has picked the home page even though a good number of internal pages, not even that deep, would be far more accurate. The home page simply isn't a good match.

Back to my gut feeling - I have the impression that I'm "allowed" to do more with internal linking as long as I have a certain number of IBL's to that page.

I haven't finished experimenting though. As said, I'm working on rephrasing some anchor text. On other affected pages I'm simply reducing the number of internal links while elsewhere I have to wait for the effect of additional IBL's.

[AbleReach]

Does age of page or lack of fresh, well-worded (?) or authoratative links leading into page seem to be a common trait?

I've seen some shifting from up there to outa there. I don't know why or if it will last, of course.

Edited by AbleReach, 11 January 2006 - 04:11 PM.

[travis]

Thats exactly what happened to one of our customer just now.

http://www.google.co...m.au ravensgate

The old pages from the previous site from 10 months ago are suddenly appearing as supplemental results, and the home page shows up as just a link.

We have done very little in terms of optimisation for these guys, and made a nice clean site in CSS, and it was fine all year.

We did make some changes to the text recently, and added a table of clients to the site. It would have represented a change of about 40% increase in text volume to the site.

That was the only change we made all year in 2005. I am sure it will fix up.

[Nadir]

Yeah that's weird Travis, your page doesn't even have cache, when was the last time Google crawled the site?

[AbleReach]

Y'know, this may be a betwixt and between thing.

Today I turned on the Google Toolbar and saw that www.google.com has 0/10 PR.
www.yahoo.com is at 9/10. Didn't look further to see what else has zero when it should have something. Stuff happens.

[JohnMu]

FWIW Travis, it looks fine from here (Switzerland), even if I use the normal dance-sites to check the other datacenters.

Just wondering if any of you use Google Sitemaps? In the Google group to Sitemaps there are more and more people that complain of similar things happening (though you can usually trace it back to something on their site).

John

[DonnaFontenot]

gotta agree with bill on the first one. dup problems are almost always the cause of pages dropping out into supplemental (at least every time i've ever checked, anyway) (and assuming the pages still exist, of course).

the second one...no idea.

[randfish]

First, to answer some questions. The site is a unique, authoritative site that is well respected in its field outside of just the web property. They've been a source of news and a big industry directory for many years - since before the web, actually. So no keyword-keyword-keyword.com

The puzzling thing is, they don't have any duplicate content - not even pseudo scraper sites pulling the page copy...

I did just notice that a few of the pages are in the index at BigDaddy, though... perhaps just a fluke?

As for the single phrase issue getting pushed far down in the results, I think I can share one of those results without issue and you can all look it in person (from my client mentioned in Newsweek - no cat in that bag):

At Google:
commercial bridge loans
hard money bridge loans
bridge loans

Here's what interesting. The sites ranking well for the combination of terms and phrases, specifically metmtge.com, avatarfinancial.com, brtrealty.com rank for the singular phrase "bridge loans" in positions 194, 331, and 71 respectively. Interestingly, too, the pages that rank well - the home pages, do not appear in the singular result (bridge loans), but do for the other searches...

I've seen this for another search result for a client I cannot share, and someone else who recently emailed me... Something new going on at Google?

Edited by randfish, 11 January 2006 - 07:49 PM.

[Nadir]

Only Google knows what's up my friend. I really don't see how anyone outside the Google sphere can explain that at this time.

[Black_Knight]

Any ideas? Anyone seen simliiar stuff?

Back when Google first changed the way they handled 301 and 302 redirects (thus enabling the 302 page jacking fiasco) this happened to one of my clients who'd just implemented server-side redirection.

The exact situation you describe was the almost instant sign.

So yes, I'd follow Bill's suggestion to look for duplicates, and I'd further suggest looking for possible 302 jacking or other redirection related issues either on the site itself, or from third parties.

[travis]

when was the last time Google crawled the site?

Nadir,

I dont really keep an eye on it, but the pages have been trawled well and truely. Its a simple site.

Its just that the supplemental results from the old site have suddenly re-appeared. I thought after 10 months, they would be dead and buried.

They are like zombie pages. They just wont die. I need a wooden stake and a silver bullet.

[egain]

We had similar goings on with one of our sites, recently however during the course of about a three week period, which as you have mentioned previously resulted in a noticable drop in both SERPs and resultant traffic.

As you have mentioned, we reviewed incoming link traffic, and much of the site architecture in order to identify any obvious trends that may be affecting the site.

To be honest, still not entirely sure what solved it, we created a Google sitemap, and apart from slightly adjusting the internal navigation structure (and other small structural tweaks), didn't do anything much different (as theyre was nothing we could put our finger on)

However, we have since recovered (signicantly) both in terms of traffic/SERPs so I can only guess it must be due to some external factors, whether they be Google/Algorithm based or other.

By the way Rand, nice work on the SEOMoz site !!!..

Edited by egain, 12 January 2006 - 10:05 AM.

[JohnMu]

...we created a Google sitemap...

I don't know if this was the case for your site or not -- but ever since the introduction of Google Sitemaps (I've been following them very closely since the beginning) I have seen isolated cases of similar things happening when a site starts using Google Sitemaps. My assumption is the following:

- You submitted a sitemap file with your full URL listing, they way you "know" them to be (be it from a crawler or from a CMS directly)

- Google notices that there is a "significant" difference between the URLs in the sitemap file and the URLs in its index. The word "significant" is the big question mark - how it notices, why it notices, why it cares.

- Google decides to clean the index for your site (remove it, more or less) and "reindex" your site to try to get a clean set of URLs for your site in the index. I presume the wait between removing the site from the index until it is reindexed with clean URLs is to allow all datacenters to catch up.

I've seen several sites that have had this happen to them, it's just a matter of time (usually a week or two) until it is cleaned up. I am not sure which parameters Google uses to determine the need to clean up the index, I assume it has something to do with the number of similar (but not identical) URLs in the sitemaps-file and the index and perhaps with the "value" of the site itself (ie low-value sites are left to get cleaned up in the long run - though it would not make sense to throw high-value sites out of the index to clean the entries (?)). Also - I don't know what it does with links to "dirty" URLs -- I assume it will either include them in the index or redirect them to a similar "clean" URL. Sitemaps is an inclusion-type listing, it is not meant to be authoritive, ie it will generally just add to the index and not remove URLs that are not included in the sitemap file.

John

[DonnaFontenot]

The puzzling thing is, they don't have any duplicate content - not even pseudo scraper sites pulling the page copy...

Not that I don't believe you, Rand , but have you checked for internal dups such as possible url problems? i.e. two different urls (with parameters, what-have-you) pointing to the same page... or print-only versions of the page... or different pages having the same meta desc / keywords ?

[AltherrWeb]

I think it's what brag and dazzlin mentioned with the dup content.
This seems to happen a lot if I pull in a php header and footer and the page isn't that long in content.
Maybe there's a certain percentage of the page that has to be unique for Google to see it as being a completely different page of information.
I would try adding more content on the page that isn't on any of the other pages and see if that helps.
It's just dumb to me, that Google wants websites that have good usability (which I interpret as the same menu in the same location on each page), and then they hold it against you when indexing.

[JohnMu]

Hi Rand
Not wanting to do your job (and not assuming I could ever do it as good :-)), but looking through your links I don't see any with just "bridge loans" as anchor text or as title of the page linking (at least not any links of value, heh). I always thought that the more you "dilute" the anchor text keywords (say "hard money consulting private bridge loans" instead of just "bridge loans") the less value the individual terms had. Perhaps Google's last update went even more in that direction? (Google doesn't list any links that have "bridge loans" in the anchor text.)

Also, just wondering, why are you serving the same content on www and non-www domains? Shouldn't you be doing a 301 redirect from one to the other?

While I'm at it, what's your reasoning behind the very short "non-loan" page titles (things like "Health Care", "Mixed-Use", "Self-Storage") and having some pages with keywords + description meta tags and others without?

I'm sure you have your reasons, I'm too new to this game to be able to understand it all, but it's fun learning more and more

John